Benchmark
WildBench: llama_reward.Information/Advice seeking.K=-1
People compare model answers and choose which they prefer. Higher scores are better.
- Models measured
- 32
- Scores recorded
- 32
- Sources
- 1
- IQ weight
- 1.67
Score distribution
- 17.5
- Mean
- 26.8
- Median
- 28.6
- Std dev
- -16.8
- 10th percentile
- 48.5
- 90th percentile
- -68.1
- Min
This benchmark has a weight of 1.67 for chat models, measured over 32 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | GPT 4oallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | 54.9 |
| 2 | Claude 3.5 Sonnetallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | 51.8 |
| 3 | GPT-4 Turboallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | 50.1 |
| 4 | Gemini 1.5 Proallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | 48.5 |
| 5 | GPT-4-0125-previewallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | 46.8 |
| 6 | Llama 3 70B Instructallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | 43 |
| 7 | Reka Coreallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | 40.2 |
| 8 | Yi 1.5 34B Chatallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | 40.1 |
| 9 | Gemini 1.5 Flashallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | 39.8 |
| 10 | Nemotron-4 340B Instructallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | 38.9 |
| 11 | Claude 3 Opusallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | 37.4 |
| 12 | yi-largeallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | 37.1 |
| 13 | DeepSeek Coder V2 Instructallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | 27.8 |
| 14 | GLM-4-9b-chatallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | 27.7 |
| 15 | Qwen2 72B Instructallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | 27.5 |
| 16 | Command R Plusallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | 26.8 |
| 17 | Claude 3 Sonnetallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | 22.1 |
| 18 | Claude 3 Haikuallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | 15.3 |
| 19 | Starling LM 7B Betaallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | 14.4 |
| 20 | Command R+allenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | 12.3 |
| 21 | Llama 3 8B Instructallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | 11.6 |
| 22 | Reka Flashallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | 9.2 |
| 23 | Mistral Largeallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf, 2402· WildBench direct data | 7.4 |
| 24 | Mixtral 8x7B Instruct v0.1allenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | 4.4 |
| 25 | Mistral 7B Instruct v0.2allenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | -0.525 |
| 26 | Reka Edgeallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | -3.7 |
| 27 | Nous Hermes 2 Mixtral 8x7B DPOallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | -6.4 |
| 28 | Tulu 2 DPO 70Ballenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | -9.2 |
| 29 | GPT-3.5 Turboallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | -16.8 |
| 30 | Llama-2-7b-chat-hfallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | -20.3 |
| 31 | Gemma 7B ITallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | -50 |
| 32 | Gemma 2B ITallenai/WildBench, llama_reward.Information/Advice seeking.K=-1, Llama-2-70b-chat-hf· WildBench direct data | -68.1 |
Data credits: WildBench, Allen Institute for AI